This competitive renewal continues to explore the general hypothesis (a) that surface expressed proteins control the interactions of endothelial cells with blood molecules, with adjacent ECs and with underlying matrix and cells of the vessel wall and tissues, and (b) that expression and organization of these surface proteins are regulated by extracellular signals such as cytokines. Surface proteins localized to the intercellular junctions of ECs control vascular permeability. The renewal will focus on the role of tight junction (TJ) proteins, especially claudin 5, in human capillary ECs, and how TJs (and claudin 5) are affected by the pro-inflammatory cytokine, tumor necrosis factor (TNF). The specific hypothesis to be tested has three components: (a) that disruption of TJs in ECs that line the continuous capillaries in sepsis and related conditions produces capillary leak, a major mechanism of organ injury and dysfunction; (b) that TNF, a major mediator of sepsis, causes capillary leak by disrupting TJs either by altering the expression or organization of proteins that directly form TJs or that control TJ organization through the actin cytoskeleton; and (c) that KLF4, a transcription factor induced in capillary ECs by laminar shear stress, can regulate the pathological effects of TNF on capillary TJs. This hypothesis will be explored in three specific aims. First, we will characterize the organization of TJs, with emphasis on the role of claudin 5, in human continuous capillary ECs and determine how TJ proteins interact with each other and with the actin cytoskeleton. Second, we will determine how TNF affects the organization of human capillary TJ proteins and identify those proteins whose expression levels are changed by TNF, resulting in TJ disruption and capillary leak. Third, we will determine how KLF4 regulates TNF-induced changes in capillary ECs and if this can be exploited to control TNF-induced pathological changes. In these experiments we will manipulate gene expression of cultured human microvascular ECs and compare with other EC systems to analyze structure and function of TJs in vitro, making use of ECIS measurements of transendothelial electrical resistance as a key functional assay. We also will examine intact human tissues and experimentally manipulate human capillaries in vivo by engrafting immuno-deficient mice with vascularized human skin or with synthetic microvascular beds constructed from wild type and genetically altered human ECs. Successful completion of these aims may lead to effective therapeutic approaches to control capillary leak.